Two-dimensional rectangular bars have been used in an experimental study to control boundary layer transition and reattachment under low-pressure turbine conditions. Cases with Reynolds numbers (Re) ranging from 25,000 to 300,000 (based on suction surface length and exit velocity) have been considered at low (0.5%) and high (8.5% inlet) free-stream turbulence levels. Three different bars were considered, with heights ranging from 0.2% to 0.7% of suction surface length. Mean and fluctuating velocity and intermittency profiles are presented and compared to results of baseline cases from a previous study. Bar performance depends on the bar height and the location of the bar trailing edge. Bars located near the suction surface velocity maximum are most effective. Large bars trip the boundary layer to turbulent and prevent separation, but create unnecessarily high losses. Somewhat smaller bars had no immediate detectable effect on the boundary layer, but introduced small disturbances which caused transition and reattachment to move upstream from their locations in the corresponding baseline case. The smaller bars were effective under both high and low free-stream turbulence conditions, indicating that the high free-stream turbulence transition is not simply a bypass transition induced by the free-stream. Losses appear to be minimized when a small separation bubble is present, so long as reattachment begins far enough upstream for the boundary layer to recover from the separation. Correlations for determining optimal bar height are presented. The bars appear to provide a simple and effective means of passive flow control. Bars which are large enough to induce reattachment at low Re, however, cause higher losses at the highest Re. Some compromise would, therefore, be needed when choosing a bar height for best overall performance.

This content is only available via PDF.
You do not currently have access to this content.